Visual perception is not always a veridical representation of our environment. Instead, processing of the visual scene is influenced by the selective allocation of attention. Vision at an attended location is faster, more accurate, and of higher spatial resolution [1-11, for a review see 12]. By improving information processing at attended locations at the cost of unattended locations, attention allows us to make most efficient use of the visual system's limited resources. Recent studies show that attention not only improves performance but also alters the appearance of several basic visual features [13-22]. Some authors have criticized the interpretation of these findings and suggested alternative explanations based on a decisional bias, i.e., people are more likely to select an attended stimulus but do not perceive it differently [23-26]. Although this account has been counter-argued by control experiments [13, 15-21] and empirical replies [14, 27, 28], the issue is still hotly debated. The physiological basis of such an attentional modulation of phenomenological experience is not yet known. Whereas an enhanced neuronal response explains increased performance with attention, it does not necessarily imply a change of appearance. Instead, the appearance of stimulus features has often been associated with the activity of neuronal populations selectively tuned for these features, and misperceptions have been related to shifts in their selectivity [49]. Therefore, a change of appearance should be mediated by shifts in the selectivity of those neuronal populations involved in processing the attended stimulus. The proposed research will be the first to test effects of spatial attention on neuronal population selectivity for basic visual features and will thereby significantly contribute to understanding the mechanisms of spatial attention. Furthermore, it will provide a definitive answer to the question whether attention truly modulates appearance. An adaptation paradigm will be employed in which neuronal populations tuned to specific stimulus features can be selectively targeted in a psychophysical paradigm, so that an interaction of adaptation and attention will reveal if these neurons mediate the attentional effect on appearance. Then, population selectivity will be directly measured with and without attention using adaptation in a functional magnetic resonance imaging (fMRI) paradigm [29]. Speed and spatial frequency were chosen as two respective examples from the dorsal and ventral streams, because attention modulates the appearance of both speed [18, 20] and spatial frequency [15, 30] and because neurons tuned for both speed and spatial frequency exist in early visual cortex [31-38]. Results from both the psychophysical and the fMRI study will provide converging and complementing evidence for how spatial attention changes appearance.